debug registers in 2.3.x

Petr Vandrovec Ing. VTEI (VANDROVE@vc.cvut.cz)
Mon, 6 Sep 1999 21:16:48 MET-1

Messages sorted by: [ date ][ thread ][ subject ][ author ]
Next message: list.kernel@iahastie.clara.net: "Remailing error, apology"
Previous message: Keith Owens: "Re: problems with Netflex 3/P (tlan)"

Hi,
I was looking into usage of ia32 debug registers in 2.3.16 kernel
and I found that anybody with ptrace rights to any process can confuse
vm86 applications, if I did not miss something, because of we are setting
DR7 lazy, but we do not check thread.debugreg[7] in do_debug() if exception
came from VM86 mode (doing handle_vm86_trap() doing in turn do_int() or
send_sig(SIGTRAP)). We check this if exception came from 32-bit code
and I think that we should do this check in all cases, not only for 32bit
code.
Is it intentional (I do not think...) or did I miss DR7 clear/load
somewhere else? Or is VMM using vm86() syscall responsible for handling
this spurious traps?
Thanks,
Petr Vandrovec
vandrove@vc.cvut.cz

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

Next message: list.kernel@iahastie.clara.net: "Remailing error, apology"
Previous message: Keith Owens: "Re: problems with Netflex 3/P (tlan)"